Research map / Label-efficient and robust learning / Recognition and long-tail
Pose-Guided Visible Part Matching for Occluded Person ReID
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add pose-guided part attention and self-mined visibility-weighted matching | PCB-based person ReID · Occluded-REID | rank-1: 59.3 -> 66.8 | Helped |
| Add pose-guided part attention and self-mined visibility-weighted matching | PCB-based person ReID · Partial-REID | rank-1: 66.3 -> 75.3 | Helped |
| Add pose-guided part attention and self-mined visibility-weighted matching | PCB-based person ReID · P-DukeMTMC-reID | rank-1: 43.6 -> 50.1 | Helped |
| Train with randomly occluded image augmentation (ablation) | PVPM · Occluded-REID | rank-1: 66.8 -> 70.4 | Helped |
| Train with randomly occluded image augmentation (ablation) | PVPM · Partial-REID | rank-1: 75.3 -> 78.3 | Helped |
| Train with randomly occluded image augmentation (ablation) | PVPM · P-DukeMTMC-reID | mAP: 29.4 -> 29.2 | Hurt |
| Add the part matching loss Lm (ablation) | PVPM · Occluded-REID | rank-1: 65.5 -> 66.8 | Helped |
| Add the part matching loss Lm (ablation) | PVPM · Partial-REID | rank-1: 73.0 -> 75.3 | Helped |
| Add the part matching loss Lm (ablation) | PVPM · P-DukeMTMC-reID | rank-1: 48.1 -> 50.1 | Helped |
| Use graph matching instead of similarity thresholding to infer pair visibility (ablation) | PVPM · Occluded-REID | rank-1: 65.1 -> 66.8 | Helped |
| Use graph matching instead of similarity thresholding to infer pair visibility (ablation) | PVPM · Partial-REID | rank-1: 71.7 -> 75.3 | Helped |
| Use graph matching instead of similarity thresholding to infer pair visibility (ablation) | PVPM · P-DukeMTMC-reID | rank-1: 48.1 -> 50.1 | Helped |
Closest papers by meaning
- Pose-Guided Feature Disentangling for Occluded Person Re-identification Based on Transformer (AAAI 2022 · similarity 0.90)
- Matching on Sets: Conquer Occluded Person Re-identification Without Alignment (AAAI 2021 · similarity 0.87)
- High-Order Information Matters: Learning Relation and Topology for Occluded Person Re-Identification (CVPR 2020 · similarity 0.86)
- Attention Disturbance and Dual-Path Constraint Network for Occluded Person Re-identification (AAAI 2024 · similarity 0.86)
- Dynamic Feature Pruning and Consolidation for Occluded Person Re-identification (AAAI 2024 · similarity 0.85)
- Occluded Person Re-identification via Saliency-Guided Patch Transfer (AAAI 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.